Microsoft Word - MBBHP Test Apparatus Tech Note_v11.docxEvaluation of Low Global Warming
Potential Refrigerants
Harrison Skye
Evaluation of Low Global Warming
Potential Refrigerants
Harrison Skye
Engineering Laboratory
National Institute of Standards and Technology
Willie E. May, Under Secretary of Commerce for Standards and Technology and Director
i
Certain commercial entities, equipment, or materials may be identified in this
document in order to describe an experimental procedure or concept adequately.
Such identification is not intended to imply recommendation or endorsement by the
National Institute of Standards and Technology, nor is it intended to imply that the
entities, materials, or equipment are necessarily the best available for the purpose.
National Institute of Standards and Technology Technical Note 1895
Natl. Inst. Stand. Technol. Tech. Note 1895, 48 pages (September 2015)
http://dx.doi.org/10.6028/NIST.TN.1895 CODEN: NTNOEF
International efforts to reduce human-induced global warming include restrictions on the
use of chemicals with a high global warming potential (GWP). The heating, ventilation, air-
conditioning, and refrigeration (HVAC&R) industry subsequently faces a phasedown of many
commonly used hydrofluorocarbon (HFC) refrigerants that have a relatively high GWP. A new
family of refrigerants known as hydrofluoroolefins (HFOs), including their mixtures with HFCs,
show promise as replacements. The overall GWP impact of refrigerant use includes both the
direct GWP related to inadvertent release of the fluid into the atmosphere, as well as an indirect
GWP caused by emissions from the power source used to energize the associated HVAC&R
equipment. In nearly all HVAC&R applications, the indirect emissions far outweigh the direct
emissions, so the efficiency of candidate replacements must be carefully quantified to guide the
selection of fluids that actually achieve a reduction in the overall GWP.
A 3.4 kW (1 ton) heat pump test apparatus has been constructed and instrumented for
measuring the cycle performance of low-GWP refrigerants; the description of that apparatus is
the focus of this report. Details are described for the system components, instrumentation, data
reduction, and uncertainty analysis. Results from baseline experiments with R134a were used to
test and verify the apparatus and data reduction procedure. The data will be used to provide a
relative comparison between the low-GWP refrigerants as quantified by metrics including
coefficient of performance and volumetric capacity.
Future tests with low-GWP refrigerants will be compared with those from baseline
refrigerants R134a and R410A. Additionally, the test data will be used to verify a new NIST
heat pump modeling tool, CYCLE_D-HX, which captures both thermodynamic and heat transfer
processes in HVAC&R equipment. Refrigerants, refrigerant mixtures, and cycle configurations
that are difficult and/or time consuming to test can be rapidly explored using the verified model.
iii
Acknowledgements
This apparatus was constructed by John Wamsley based on the initial design by David
Yashar. Dr. Yashar also helped create the outline of the report, which was valuable for keeping
the presentation of information organized and complete. I would like to acknowledge Vance
Payne for his help in determining the operating procedures and target operating parameters.
Patrick Goenner refined the instrumentation and collected early sets of data while visiting as a
guest researcher from Germany. I would also like to thank Piotr Domanski and Riccardo
Brignoli for reviewing data sets and working on the CYCLE_D-HX model. Thanks to Mark
Kedzierski for thoughtful insight into the calibration and uncertainty analysis for the test rig.
Tara Fortin, of the NIST Applied Chemicals and Materials Division, performed differential
scanning calorimeter measurements of the heat transfer fluid specific heat; these data were
critical for achieving an acceptable energy balance. Finally, I would like to thank the reviewers
for their helpful suggestions, including Piotr Domanski and Amanda Pertzborn at NIST, and
Steve Brown at The Catholic University of America.
iv
2.3.2.1 Operating Parameters 1 & 2: Evaporator and Condenser Saturation
Temperature ...................................................................................................... 15
2.3.2.2 Operating Parameters 3, 4, & 5: Capacity and Heat Flux ................................ 17
2.3.2.3 Operating Parameters 6 & 7: HTF-side Thermal Resistance Ratio ................. 18
2.3.2.4 Operating Parameters 8 & 9: Subcool and Superheat ...................................... 18
2.3.2.5 Operating Parameter 10: LLSL-HX included or bypassed .............................. 19
3 Data Analysis .......................................................................................................................... 20
3.1 Thermodynamic States .................................................................................................... 20
3.2 Thermodynamic Performance ......................................................................................... 20
3.2.4 Volumetric Capacity ................................................................................................ 22
5 Conclusions and Future Work ................................................................................................ 28
6 References ............................................................................................................................... 29
A.1 Symbols Used in Uncertainty Analysis ........................................................................... 31
A.2 General Remarks ............................................................................................................. 32
A.4 Thermocouples with Ice-Water Bath Compensation ...................................................... 33
A.5 Thermopiles in Heat Transfer Fluid ................................................................................ 34
A.6 Evaporator Heat Transfer Fluid Specific Heat ................................................................ 36
A.7 Condenser Heat Transfer Fluid Specific Heat ................................................................. 37
A.8 Evaporator and Condenser Insulation Conductance ....................................................... 37
A.9 Uncertainty Analysis Software ........................................................................................ 39
A.10 Moving Window and Steady State Uncertainty .............................................................. 40
A.11 Uncertainty Results Summary ......................................................................................... 41
A.11.1 Uncertainty Results Summary: Cooling .................................................................. 41
A.11.2 Uncertainty Results Summary: Heating................................................................... 43
Appendix C : Microfin Tube Surface Area ................................................................................... 48
vi
Figure 2-2: Pictures of MBBHP apparatus ................................................................................ 6
Figure 2-3: Pictures of MBBHP compressor (a) without and (b) with safety cage. .................. 7
Figure 2-4: Schematic showing the tube numbering convention in the (a) condenser and (b)
evaporator. .............................................................................................................. 8
Figure 2-5: Schematics of annular heat exchanger including (a) refrigerant tube lengths, (b)
cross section of annular heat exchanger, (c) detail cross-section of microfin tube,
and (d) helix angle of microfins. ............................................................................. 9
Figure 2-6: Capacity and heat flux for baseline R134a tests for (a) evaporator and (b)
condenser .............................................................................................................. 18
Figure 3-1: Temperature profile in (a) evaporator, showing superheat and (b) condenser,
showing subcooling .............................................................................................. 23
Figure 4-1: Energy imbalance in the evaporator and condenser ............................................. 25
Figure 4-2: Heat pump cycle pressures on the (a) high pressure side and (b) low pressure
side ........................................................................................................................ 26
Figure 4-3: Performance metrics including (a) compressor power and (b) mass flow and mass
flux ........................................................................................................................ 26
Figure 4-4: Performance metrics including (a) COP, (b) capacity, (c) volumetric capacity, and
(d) compressor isentropic and volumetric efficiency ............................................ 27
Figure 4-5: Flow regime map for microfin tube (a) evaporation and (b) condensation .......... 28
Figure A-1: Temperature calibration for CJC compensated thermocouples where (a) data are
divided by each of the 5 calibrations and (b) all data are combined..................... 33
Figure A-2: Ice-water bath referenced thermocouple calibration data ..................................... 34
Figure A-3: Thermopile calibration data .................................................................................. 35
Figure A-4: Contours of thermopile (a) uncertainty and (b) relative contribution to
uncertainty............................................................................................................. 36
Figure A-6: Specific heat of condenser HTF (water) ............................................................... 37
Figure A-7: Insulation conductance for evaporator and condenser .......................................... 39
Figure A-8: Uncertainty in temperature due to fluctuations at steady state for (a) evaporator
outlet and (b) evaporator inlet ............................................................................... 41
vii
Table 2-2: Refrigerant microfin tube specifications.................................................................. 11
Table 2-3: List of MBBHP instruments (SI units) .................................................................... 13
Table 2-4: List of MBBHP instruments (I-P units) ................................................................... 14
Table 2-5: List of the Control Parameters and Operating Parameters ...................................... 15
Table 2-6: Evaporator and condenser fluid temperatures for standard rating tests ................... 17
Table 3-1: Measurements and equations used to define the thermodynamic states .................. 20
Table A-1: Condenser & evaporator thermopile polynomial coefficients ................................. 35
Table A-2: Evaporator HTF specific heat polynomial coefficients ........................................... 37
Table A-3: Uncertainty of refrigerant-side evaporator cooling capacity ................................... 42
Table A-4: Uncertainty of HTF-side evaporator cooling capacity ............................................ 42
Table A-5: Uncertainty of cooling COP .................................................................................... 43
Table A-6: Uncertainty of refrigerant-side condenser heating capacity .................................... 44
Table A-7: Uncertainty of HTF-side condenser heating capacity.............................................. 44
Table A-8: Uncertainty of heating COP ..................................................................................... 45
viii
Nomenclature
Symbol Units Definition
A m2 Area
Atotal m2 Total combined heat transfer area from evaporator and condenser
COPheat -- Coefficient of Performance - Heating
COPcool -- Coefficient of Performance - Cooling
Cp kJ kg-1 K-1 Specific heat
do mm Microfin tube outer diameter
di mm Microfin tube inner diameter
Dcomp m3 Compressor displacement
L m Tube length
N Hz Compressor frequency
Nf -- Number of microfins on inner circumference of refrigerant tube
NT -- Number of heat exchanger tubes
P kPa Pressure
R K kW-1 Thermal resistance
Rtotal K kW-1 Total resistance in heat exchanger including both fluids and tube wall
Q kW Energy transfer
,LLSL vQ kW Energy transfer in LLSL-HX, on the vapor side
,LLSL maxQ kW Maximum possible energy transfer in LLSL-HX
T °C Temperature
sf mm Spacing between microfins
s kJ kg-1 K-1 Specific entropy
SC K Condenser subcooling
SH K Evaporator superheat
VCC kJ m-3 Volumetric Cooling Capacity
VHC kJ m-3 Volumetric Heating Capacity
compW kW Compressor power, computed using enthalpy change of refrigerant
shaftW kW Compressor power, computed using torque and speed of driveshaft
x -- Vapor quality (mass of vapor divided by total mass of fluid)
ix
Greek
β ° Microfin angle
ηs -- Compressor isentropic efficiency
ηv -- Compressor volumetric efficiency
τ N m Compressor shaft torque
Subscript Definition
air Air
avg Average
c Condenser
e Evaporator
out Outlet
ref Refrigerant
1 to 11 Refrigerant thermodynamic states as defined in Figure 2-1
x
ANSI American National Standards Institute
CFC Chlorofluorocarbon
COP Coefficient of Performance (W of capacity per W of electric input)
CYCLE_D-HX A NIST heat pump cycle simulation model currently under development.
The simulation captures both thermodynamic and transport processes in the
cycle.
DOE Department of Energy (United States)
EER Energy Efficiency Ratio (Btu of capacity per W of electric input)
EOS Equation of State
GWP Global Warming Potential
LLSL-HX Liquid-Line Suction-Line Heat Exchanger
MBBHP Mini Breadboard Heat Pump
NIST National Institute of Standards and Technology (United States)
PID Proportional Integral Derivative controller
RPM Revolutions per Minute (compressor shaft)
1
1 Introduction
Concerns about the environmental impacts of global warming (i.e., global climate change)
are driving an effort to limit anthropogenic sources of atmospheric pollutants that trap longwave
radiation emitted from the earth’s surface. The chlorinated- and fluorinated-hydrocarbon
working fluids (i.e., refrigerants) employed by the heating, ventilation, air-conditioning, and
refrigeration (HVAC&R) industry exhibit a particularly large global warming potential (GWP)
with 100-year GWP values hundreds to thousands of times larger than an equivalent mass of
carbon dioxide (Solomon, et al., 2007, p. 212). In the European Union, the F-gas regulation (EU,
2014) mandates a phase-down of hydrofluorocarbons (HFCs) with high GWP to 21 % of the
average levels from years 2009 through 2012 by 2030. The current North American proposal
would amend the Montreal Protocol to limit HFC consumption to have a weighted GWP of 15 %
of average levels from years 2011 through 2013 by 2036 (EPA, 2015).
Major efforts are underway to identify alternative refrigerants with a lower GWP. Chemical
manufacturers are proposing halogenated olefins (e.g., hydrofluoroolefins, or HFOs), which offer
substantially lower GWP due to short atmospheric lifetimes. For some HFOs, the lower GWP
comes at the cost of an increase in flammability. The Air-Conditioning, Heating, and
Refrigeration Institute (AHRI) is leading a collaborative effort to evaluate the drop-in and soft-
optimized performance of low-GWP alternatives largely consisting of HFOs and HFO/HFC
blends (AHRI, 2015). The National Institute of Standards and Technology (NIST) investigated
the maximum thermodynamic potential expected for working fluids by optimizing the
parameters governing the Equations of State (EOS) (Domanski et al., 2014). In a companion
NIST study, a coarse filter criteria was developed (based on the atomic elements, GWP, toxicity,
flammability, critical temperature, and stability) to sift through the 100 million chemicals
currently listed in the public-domain PubChem database for possible low-GWP refrigerant
candidates, 1234 of which were identified (Kazakov et al., 2012). The candidate list was further
refined to 62 by limiting the critical temperature to the range 300 K to 400 K, which is typical for
fluids in HVAC&R equipment (McLinden et al., 2014).
The remaining 62 candidates, and their mixtures, must be evaluated using more stringent
criteria that consider detailed performance in HVAC&R equipment. In particular, the criteria
must include the cycle efficiency of the fluid since the carbon dioxide emissions from the power
source (i.e., “indirect emissions”) far outweighs the direct GWP impact (i.e., the GWP from the
release of the fluid into the atmosphere) of the working fluid in nearly all HVAC&R equipment.
A cycle modeling tool, CYCLE_D-HX, is being developed at NIST to evaluate the efficiency of
low-GWP refrigerants and refrigerant mixtures. The model expands upon the thermodynamic
considerations from CYCLE_D (Brown et al., 2011) by including transport phenomena. In this
way, appropriate penalties can be applied to refrigerants whose thermodynamic properties are
favorable but exhibit poor performance in HVAC&R systems because of large pressure drops or
poor heat transfer. The consideration of transport processes represents a lesson learned from the
previous effort to find replacements for chlorofluorocarbon (CFC) fluids that were found to
cause stratospheric ozone destruction. R410A, the dominant replacement for the CFC R22, was
initially considered to be a significantly inferior replacement based on thermodynamic properties
alone. However, later practice showed that the lower pressure drop (due to high operating
pressure) and superior heat transfer of R410A could be exploited with optimized heat exchangers
to yield performance competitive with R22. Including transport processes also allows for the
2
optimization of mass flux; the number of parallel heat exchanger tube circuits are adjusted to
balance performance enhancement/degradation of increased heat transfer/pressure drop with
higher mass flux. The more sophisticated capabilities of the new model will allow for more
accurate identification of optimal low-GWP fluids using computational methods.
A heat pump test apparatus has been constructed for the purpose of generating cycle
performance data to compare low-GWP refrigerants and to tune and verify the new NIST model;
that test apparatus is the focus of this report. The apparatus is highly configurable, which allows
the system to operate with a wide variety of working fluids as if tailored to each one.
Specifically, the apparatus features a variable-speed compressor, variable-area heat exchangers,
and a manually adjusted throttle valve. These adjustable components allow for operation at
constant capacity per total combined evaporator and condenser conductance ( )totalQ UA across
all test fluids, a condition required for equitable comparison of refrigerants (as well as fixed heat
transfer fluid inlet/outlet temperatures) (McLinden et al., 1987). In practice, it is difficult to hold
totalQ UA constant because the conductance depends on the refrigerant side heat transfer
coefficient, which changes for each fluid and operating condition. Rather, a fixed capacity per
total heat exchanger area ( totalQ A ) is used as a close approximation. For example, with a fixed
heat exchanger area, totalQ A is held constant from R134a (low volumetric capacity) to R410A
(high volumetric capacity) by decreasing the compressor speed. If adjustment of compressor
speed is insufficient to maintain constant totalQ A between refrigerants, the heat exchanger area
can be adjusted as well. The adjustability and relatively small capacity of 1.7 kW to 3.5 kW (0.5
ton to 1 ton) provide the etymology for the apparatus name, the Mini Breadboard Heat Pump
(MBBHP). Measurements of cycle performance with legacy high-GWP HFCs and novel low-
GWP fluids in the MBBHP will provide a rich data set to test the predictive capability of the new
NIST model. This report describes the construction and operation of the test apparatus, the data analysis
used to quantify performance, an uncertainty analysis of the performance metrics, and baseline
and validation results using refrigerant R134a.
3
2.1 Components
The test apparatus, shown in Figure 2-1 through Figure 2-3, was designed to emulate
refrigerant-side operating conditions of typical HVAC&R equipment. (Figure B-1 and Figure B-
2 in Appendix B show the instrument numbering convention used in the raw data files). The rig
design is partially based on one used in a previous NIST effort to investigate alternatives to
CFCs (Pannock et al., 1991). The refrigerant circuit includes a variable-speed compressor
powered by an electric motor and inverter, where the speed controls cooling/heating capacity.
The evaporator and condenser, shown in detail in Figure 2-4 and Figure 2-5 (a) and (b), are
single-circuit (i.e., there are no parallel tube branches) annular heat exchangers where the
refrigerant is on the inner tube and the liquid heat transfer fluid (HTF) is in the annular space.
As shown in Figure 2-5(c), the refrigerant tube is smooth on the outside and enhanced with rifled
microfins on the inside. The size of the heat exchangers can be adjusted by changing the number
of active refrigerant tubes; the valves on the side of the heat exchanger allow for up to 10 (in
increments of 2) of the 20 refrigerant tubes to be bypassed (this section of the rig is referred to as
the “bypass valve header”). When operating these bypass valves, care is taken to avoid trapping
subcooled liquid in a closed section; a section completely filled with subcooled liquid will
experience a catastrophic increase in pressure if the temperature rises enough to vaporize the
refrigerant. A liquid-line suction-line heat exchanger (LLSL-HX) can be included or bypassed
as needed. Each of three heat exchangers (evaporator, condenser and LLSL-HX) are arranged in
counterflow configuration. Two sizes of manually controlled Vernier-handled needle valves are
used to throttle the refrigerant and can be used individually or in tandem; the needle valve sizes
were selected using a correlation for short tubes (Kim et al., 2005). A filter-dryer is used to
control contaminants, and an oil separator is used to ensure proper oil return to the compressor
and to minimize oil circulation in the heat exchangers. The system refrigerant charge is adjusted
using two Shrader access valves, where refrigerant is removed from or added to the
liquid/suction line valves, respectively. Two 40 W heaters preheat the compressor before startup
to avoid foam in the oil (the heaters are turned off after the discharge temperature has warmed to
about 60 °C). A safety cage was constructed around the compressor because during cooling tests
with high pressure fluids such as R410A, the compressor will be operated slightly above the
manufacturer’s specified maximum pressure of 2600 kPa (363 psig). Finally, a series of sight
glasses provide visual confirmation of fully subcooled liquid exiting the condenser, vapor only
(i.e., no liquid) entering the compressor suction port, proper oil level in the compressor, and
proper oil return from the oil separator. Detailed component specifications are listed in Table
2-1. The microfin tube parameters are shown in Figure 2-5 and Table 2-2, and Appendix C
shows how the microfin surface area was calculated.
The HTF circuits include a water heater/chiller to respectively control the evaporator and
condenser HTF inlet temperatures, where the heater/chiller contains a pump for circulating the
fluid. The evaporator HTF is a potassium formate and water solution with freeze protection
to -40 °C (-40 °F), whereas water is used for the HTF in the condenser. Flow control provided
by three valves in each HTF circuit allows for a wide range of flowrates: two in-line valves are
used where the larger/smaller (gate/needle) valve provides gross/fine flow adjustment and the
third valve bypasses the heater (or chiller) for flow and pressure control. Air bubbles in the
condenser HTF circuit are controlled by drawing from the bottom of the vented chiller reservoir
4
tank. The evaporator HTF circuit is pressurized with an expansion tank, so air bubbles can be
removed by venting trapped air from local high spots using Shrader valves. Experience
operating the test apparatus has shown that the vented reservoir tank (in the chiller) is a superior
solution for minimizing air bubbles. The temperature of the fluid delivered to the heat
exchangers is regulated using proportional-integral-derivative (PID) controllers.
5
6
7
(a) (b)
Figure 2-3: Pictures of MBBHP compressor (a) without and (b) with safety cage.
8
(a) (b)
Figure 2-4: Schematic showing the tube numbering convention in the (a) condenser
and (b) evaporator.
(b) (c) (d)
Figure 2-5: Schematics of annular heat exchanger including (a) refrigerant tube
lengths, (b) cross section of annular heat exchanger, (c) detail cross-
section of microfin tube, and (d) helix angle of microfins.
10
Component Characteristics
Compressor Type: Reciprocating
(500 RPM to 4000 RPM)
Displacement rate: 1.31 m3h-1 to 10.81 m3h-1
Displacement: 7.165 cm3
Electric motor Maximum power: 3.73 kW (5 hp)
Expansion tank (HTF) Size: 19 liter (5 gallon)
Filter (HTF) Filter size: 50 µm
Filter-dryer (refrig.) Line size: 9.5 mm (0.375 in)
Inverter Voltage: 230 VAC, 3-phase
Maximum power: 3.73 kW (5 hp)
Heater Heating capacity: 10 kW (2.8 ton)
Heat exchanger
Tube active length: 560 mm (22 in)
Return bend length: 83 mm (3.25 in)
Outer tube ID: 25.4 mm (1 in)
Outer tube OD: 26.8 mm (1.125 in)
Outer tube material: Copper
Liquid-line suction-line heat
Type: Corrugated (liquid side)
Orifice:
Flow coeff. (Cv): 0.73*
Height: 356 mm (14 in)
Capacity: 10 to 17 kW (3 to 5 ton)
Pressure relief valve Cracking pressure: 800 psig
*Note: valve orifice sizes and flow coefficients (Cv) can vary between 0 and the listed
maximum values
Parameter Symbol Value Uncertainty
Inside diameter di 8.46 mm (0.333 in) Unknown
Outside diameter do 9.52 mm (0.375 in) ±0.050 mm (±0.002 in)
Bottom wall thickness tb 0.33 mm (0.013 in) ±0.025 mm (±0.001 in)
Fin height e 0.20 mm (0.008 in) ±0.025 mm (±0.001 in)
Spacing between fins sf 0.22 mm (0.009 in) Unknown
Number of fins Nf 60 --
Fin angle β 60° Unknown
Helix angle α 18° 2°
Internal surface area
Material Copper
2.2 Instrumentation
Refrigerant-side instrumentation is selected to capture the key thermodynamic state points in
the cycle. Specifications for the instruments (including instrument uncertainty) in both the
refrigerant and HTF circuits are listed in Table 2-3 (Table 2-4 shows I-P units). The pressure
transducers and in-stream thermocouple probes are numbered in Figure 2-1 according to the
eleven states computed in Section 3.1; the thermodynamic state number convention is also
consistent with that used in the new CYCLE_D-HX model. The thermocouple probes are
referenced to a cold junction in an ice-water bath. Differential pressure transducers quantify the
pressure drop in the condenser, evaporator, and suction line. The location of the heat exchanger
pressure taps has been selected at the bottom of the bypass valve header, where the pressure is
equal to that of the refrigerant leaving the heat exchanger regardless of how many tubes are
bypassed. A coriolis flowmeter measures the refrigerant mass flow and density, and provides
confirmation of fully subcooled liquid in the liquid line (a two-phase mixture will exhibit large
oscillations in measured flowrate). The compressor shaft power is captured by the in-line torque
meter and tachometer.
On the HTF fluid side, energy transfers are measured in the evaporator and condenser to
provide verification of the refrigerant-side calculations. The mass flow and temperature
difference in the heat exchangers are measured with coriolis meters and thermopiles (in
thermowells), which are combined with the fluid heat capacity to compute energy transfer.
Thermocouples inserted into the thermowells with the thermopiles provide the measure of
absolute temperature required for applying the thermopile calibration. Additionally, the
thermocouples provide verification of the thermopile temperature difference measurement.
Figure 2-1 and Figure 2-4 also show the tube-surface-mounted thermocouples used to
measure both refrigerant and HTF temperature profiles in the evaporator and condenser. As
shown in Section 3.2.6, these temperature profiles are useful for determining refrigerant tube
12
number where the phase transitions (superheat, subcooling, two-phase) occur. The refrigerant-
and HTF-side sensors are mounted on the return bends of each inner/outer tube, respectively.
The instruments are scanned once per minute by the data acquisition (DAQ) system, and
when steady state is achieved, the measurements are calculated as the average of a moving
window of 30 readings (i.e., the average over the last 30 minutes). The measurements inevitably
fluctuate during steady state, which contributes to the uncertainty of the measurement. This
“steady state” uncertainty has been quantified for a nominal test condition and is listed in Table
2-3. The total measurement uncertainty (last column of Table 2-3) is found by adding the
instrument error and the steady state error in quadrature. A detailed discussion of the selection
of the steady state moving window size and computation of steady state uncertainty is shown in
Appendix A.10.
Instrument Range
Density:
Mass flow 3:
Density:
Mass flow 4:
±0.0005 gcm-3
Data acquisition system (-10 to 10) V ±1 µV -- ±1 µV
Press. transducer (diff.) – condenser (0 to 172) kPa ±0.7 kPa ±0.8 kPa ±1.3 kPa
Press. transducer (diff.) – evaporator (0 to 345) kPa ±0.1 kPa ±0.8 kPa ±0.8 kPa
Press. transducer (diff.) – suction (0 to 69) kPa ±0.01 kPa ±0.3 kPa ±0.3 kPa
Press. transducer (high press.) (0 to 3447) kPa ±3.4 kPa ±0.7 kPa ±3.5 kPa
Press. transducer (low press.) (0 to 1724) kPa ±3.4 kPa ±0.7 kPa ±3.5 kPa
Tachometer (0.17 to 1667) Hz
(10 to 99 999) RPM
±0.0167 Hz
±1 RPM
±0.005 Hz
±0.3 RPM
±0.0173 Hz
±1.04 RPM
Thermocouple (in-stream probe, ice ref.) (-10 to 100) °C ±0.08 °C ±0.06 °C ±0.09 °C
Thermocouple (surface-mount, CJC) (-10 to 60) °C ±0.6 °C ±0.04 °C ±0.6 °C
Thermopile5 (1 to 10) K ±0.015 K ±0.013 K ±0.0153 K
Thermopile voltage (-10 to 10) V ±1 µV ±6.5 µV ±7 µV
Torque meter (0 to 40) N m ±0.33 Nm ±0.005 Nm 0.33 Nm
1 95% confidence interval (k = 2) 2 % rdg values have been applied to nominal values 3 Maximum of 0.1% and (z.s./flowrate)*100 %, where z.s. is zero stability = 0.0491 g s -1. 4 Maximum of 0.1% and (z.s./flowrate)*100 %, where z.s. is zero stability = 0.0075 g s -1. 5 For temperature differences in 1 K to 10 K range. Outside range, uncertainty is higher. See Appendix A.5 for more
detail.
14
Instrument Range
Density:
Mass flow 3:
Density:
Mass flow 4:
±0.062 lb ft-3
Data acquisition system (-10 to 10) V ±1 µV -- ±1 µV
Press. transducer (diff.) – condenser (0 to 25) psi ±0.1 psi ±0.1 psi ±0.18 psi
Press. transducer (diff.) – evaporator (0 to 50) psi ±0.015 psi ±0.1 psi ±0.1 psi
Press. transducer (diff.) – suction (0 to 10) psi ±0.0015 psi ±0.04 psi ±0.04 psi
Press. transducer (high press.) (0 to 500) psia ±0.5 psi ±0.1 psi ±0.51 psi
Press. transducer (low press.) (0 to 250) psia ±0.5 psi ±0.1 psi ±0.51 psi
Tachometer (0.17 to 1667) Hz
(10 to 99 999) RPM
±0.0167 Hz
±1 RPM
±0.005 Hz
±0.3 RPM
±0.0173 Hz
±1.04 RPM
Thermocouple (in-stream probe, ice ref.) (14 to 212) °F ±0.15 °F ±0.11 °F ±0.18 °F
Thermocouple (surface-mount, CJC) (14 to 140) °F ±1.1 °F ±0.08 °F ±1.1 °F
Thermopile5 (1.8 to 18) °F ±0.027 °F ±0.023 °F ±0.04 °F
Torque meter (0 to 350) in lb ±2.9 in lb ±0.045 in lb ±2.9 in lb
1 95% confidence interval (k = 2) 2 % rdg values have been applied to nominal values 3 Maximum of 0.1% and (z.s./flowrate)*100 %, where z.s. is zero stability = 0.0065 lb min-1 4 Maximum of 0.1% and (z.s./flowrate)*100 %, where z.s. is zero stability = 0.001 lb min-1 5 For temperature differences in 1.8 °F to 18 °F range. Outside range, uncertainty is higher. See Appendix A.5 for more
detail.
15
There are ten Control Parameters (i.e., settings of adjustable components) listed in Table 2-5
that the MBBHP operator sets to achieve a desired set of Operating Parameters. Each Control
Parameter primarily governs one or two Operating Parameters; the correspondence is shown in
the table.
Table 2-5: List of the Control Parameters and Operating Parameters
# Operating Parameter Operating Parameter Values Control Parameter
1 Evaporator saturation temperature See Table 2-6 Heater setpoint (HTF inlet temp.)
2 Condenser saturation temperature See Table 2-6 Chiller setpoint (HTF inlet temp.)
3 Capacity/heat flux 1760 W to 3520 W
(0.5 ton to 1 ton)
Compressor speed
4 Evaporator heat flux 2000 W m-2 to 12 000 W m-2
(600 Btu h-1 ft-2 to 3800 Btu h-1 ft-2)
Number of evaporator tubes
5 Condenser heat flux 2000 W m-2 to 12 000 W m-2
(600 Btu h-1 ft-2 to 3800 Btu h-1 ft-2)
Number of condenser tubes
6 Condenser HTF-side resistance 50 % to 70 % of overall resistance Condenser HTF flowrate
7 Evaporator HTF-side resistance 70 % to 85 % of overall resistance Evaporator HTF flowrate
8 Subcool/ Superheat ≤ 2 tubes1 / ≤ 1 tube2 Refrigerant needle valve setting
9 Subcool/ Superheat ≤ 2 tubes1 / ≤ 1 tube2 Refrigerant charge
10 LLSL-HX included/bypassed Included/bypassed LLSL-HX valves
1achieved with subcooling of 2 K to 3 K (1.8 °F to 3.6 °F) 2achieved with superheat of 3 K to 6 K (5.4 °F to 10.8 °F)
2.3.2 Test Rig Operating Parameters
This section describes each of the target Operating Parameters, and how they are set using
the Control Parameters. Note that the headings for Sections 2.3.2.1 through 2.3.2.5 include the
Operating Parameters numbers (1 to 10) as listed in Table 2-5.
2.3.2.1 Operating Parameters 1 & 2: Evaporator and Condenser Saturation Temperature
The U.S. Department of Energy (DOE) requires heat pumps and air conditioners to be rated
at conditions prescribed in the ANSI/AHRI 210/240-2008 (AHRI, 2008) standard. The standard
specifies temperatures of the HTF (air, in the standard) entering the evaporator and condenser. A
somewhat different, but derivative, method is employed here; the HTF inlet temperatures are set
to achieve refrigerant saturation temperatures expected under ANSI/AHRI Cooling A, Cooling B
and Heating H1 Rating Tests for the baseline refrigerant R134a.
The target saturation temperatures are computed first for the Cooling A test where the
assumed Coefficient of Performance (COPcool) is 4.4 (energy efficiency ratio, EER, is 15). The
16
ANSI/AHRI standard is not prescriptive about airflow, so typical values per unit evaporator
capacity are used: for the evaporator ,( )air evap evapV Q , 0.189 m3 s-1kW-1 (400 ft3 min-1ton-1), and
for the condenser ,( )air cond evapV Q , 0.378 m3 s-1kW-1 (800 ft3 min-1ton-1). Inlet air temperatures
(Tair,in) prescribed by the ANSI/AHRI standard are listed in Table 2-6; the temperature change
across the evaporator and condenser (Tair,evap, Tair,cond) is 12.6 K and 9.7 K (22.8 °F and 17.5
°F), respectively, as computed using an energy balance on the airstream, where the condenser
energy is larger by an amount that reflects the compressor power (in the COP term):
( ) 1
air p air
cool air p air
T V Q Cρ
,air cond evapV Q = condenser airflow per unit evaporator capacity
,air evap evapV Q = evaporator airflow per unit evaporator capacity
ρair = density of air
The air outlet temperatures (Tair,out) are then computed by respectively subtracting or adding the
Tair,evap, Tair,cond, from the Tair,in values.
The average temperature difference between the refrigerant and the air (Tair,ref) for
Cooling A is specified as 10 K and 3.3 K (18 °F and 6 °F) in the evaporator and condenser,
respectively, based on experience and measurements at NIST for air-source heat pumps with comparable COPs. Table 2-6 shows the average air (Tair,avg) and average refrigerant saturation
temperatures (Tref,avg) computed using this method. The Cooling B refrigerant saturation
temperatures are computed using the same Tair and Tair,ref. In heating, the heat exchangers
operating as the evaporator/condenser swap, so the values of Tair and Tair,ref for the evaporator
(in cooling) are used for the condenser, and vice versa. The resulting refrigerant saturation
temperatures are nominally aligned with values observed in heat pump measurements at NIST.
The HTF inlet temperatures (THTF,in) are adjusted to achieve the target refrigeration saturation
temperature at a compressor speed of 30 Hz (1800 RPM) with R134a, and then held constant for
all other speeds. The corresponding HTF outlet temperatures (THTF,out) are defined as:
, , , ,HTF e out HTF e in eT T T= + (2.1)
, , , ,HTF c out HTF c in cT T T= + (2.2)
where Te and Tc are the HTF temperature differences across the evaporator/condenser
measured by the thermopiles. The THTF,out values differ for each speed; Table 2-6 shows the
values at 30 Hz (1800 RPM) as a nominal example. As discussed in Section 1, a completely fair
comparison of refrigerants requires the same HTF inlet and outlet temperatures (McLinden et al.,
1987). However, it is only possible to match either the evaporator or condenser (but not both)
THTF,out by adjusting the heat pump capacity (via compressor speed); the other THTF,out will be an
17
uncontrolled value, which will vary somewhat from the R134a value, depending on the COP of
each fluid (note the HTF mass flow rate is fixed as described in Section 2.3.2.3, so it cannot be
adjusted to control the other THTF,out). Tests with future refrigerants will be carried out with an
evaporator or condenser THTF,out equal to the measurements for R134a in cooling or heating,
respectively, and the other THTF,out will vary for each fluid.
Table 2-6: Evaporator and condenser fluid temperatures for standard rating tests
Rating Test
Tair,in Tair Tair,out Tair,avg Tair,ref Tref,avg THTF,in THTF,out 1 HTFm
°C K °C °C K °C °C °C kg s-1
(°F) (°F) (°F) (°F) (°F) (°F) (°F) (°F) lb min-1
Condenser
Cooling A 35.0 9.7 44.7 39.9 3.3 43.2 33.9 39.1 0.098
(95.0) (17.5) (112.5) (103.8) (6.0) (109.8) (93.0) (102.4) (13)
Cooling B 27.8 9.7 37.5 32.6 3.3 36.0 24.9 30.9 0.098
(82.0) (17.5) (99.5) (90.8) (6.0) (96.8) (76.7) (87.6) (13)
Heating H1 21.1 12.6 33.8 27.4 10.0 37.4 32.0 36.3 0.098
(70.0) (22.8) (92.8) (81.4) (18.0) (99.4) (89.5) (97.3) (13)
Evaporator
Cooling A 26.7 12.6 14.0 20.3 10.0 10.3 20.2 15.4 0.131
(80.0) (22.8) (57.3) (68.6) (18.0) (50.6) (68.5) (59.7) (17.3)
Cooling B 26.7 12.6 14.0 20.3 10.0 10.3 19.9 14.2 0.131
(80.0) (22.8) (57.3) (68.6) (18.0) (50.6) (67.8) (57.6) (17.3)
Heating H1 8.3 9.7 -1.4 3.5 3.3 0.1 10.3 6.5 0.131
(47.0) (17.5) (29.5) (38.3) (6.0) (32.3) (50.5) (43.7) (17.3)
1 Values for R134a at a compressor speed of 30 Hz (1800 RPM)
2.3.2.2 Operating Parameters 3, 4, & 5: Capacity and Heat Flux
Baseline capacities were established for R134a at speeds of 23.3 Hz to 36.7 Hz (1400 RPM to
2200 RPM), in 3.3 Hz (200 RPM) increments, for each Rating Test. Tests with subsequent
refrigerants will be carried out with compressor speeds that match the capacities from the
baseline tests; this ensures the refrigerants are tested at the same totalQ A . The target refrigerant-
side heat flux range was 2 kW m-2 to 12 kW m-2 (600 Btu h-1 ft-2 to 3800 Btu h-1 ft-2) to coincide
with values from typical HVAC&R equipment. The average heat flux in the evaporator and
condenser was controlled by setting the number of active tubes to 10 and 12, respectively.
Figure 2-6 summarizes the capacity for the baseline tests, where the evaporator capacity varies
from 1.2 kW to 2.1 kW (0.34 ton to 0.60 ton) and condenser capacity ranges from 1.6 kW to 2.8
kW (0.45 ton to 0.80 ton). The heat flux in the evaporator varies from 5 kW m-2 to 9 kW m-2
(1600 Btu h-1 ft-2 to 2900 Btu h-1 ft-2) and in the condenser from 5 kW m-2 to 10 kW m-2 (1700
Btu h-1 ft-2 to 3200 Btu h-1 ft-2), meeting the target heat flux range.
18
(a) (b)
Figure 2-6: Capacity and heat flux for baseline R134a tests for (a) evaporator and (b)
condenser
2.3.2.3 Operating Parameters 6 & 7: HTF-side Thermal Resistance Ratio
In typical air-to-air heat pumps the air-side convection provides the majority of the
resistance to heat transfer. Evaporation heat transfer coefficients are generally larger than those
for condensation, so the HTF-side resistance ratio (RHTF/Rtotal) is selected in the range of 50 % to
70 % for the condenser and 70 % to 85 % in the evaporator. The CYCLE_D-HX model reports
heat exchanger thermal resistances for both the evaporator and condenser when given
measurements of HTF in/out temperatures, capacity, compressor efficiency, and heat exchanger
geometry, pressure drop, and log-mean temperature difference. These resistances include the
heat exchanger total conductance (UA) (the total resistance, Rtotal, is the inverse of UA), the heat
transfer-side resistance (RHTF) in the evaporator/condenser, and the refrigerant-side thermal
resistance (Rref) in the evaporator/condenser. Note that the tube wall thermal resistance is
neglected.
The MBBHP was operated under varied HTF flowrates ( HTFm ), and the resulting data were
processed using the CYCLE_D-HX model to determine RHTF/Rtotal for each flowrate. The
evaporator and condenser HTF flowrates used for the tests are 0.131 kgs-1 and 0.098 kgs-1 (17.3
lb min-1 and 13 lb min-1) as shown in Table 2-6, which yield RHTF/Rtotal of 80 % and 60 % for the
baseline Cooling A test with R134a at 30 Hz (1800 RPM). The flowrates are kept constant for
all other speeds, test conditions, and refrigerants. Maintaining a constant flowrate is important
for verifying the CYCLE_D-HX model; the model evaluates the HTF-side thermal resistance for
a reference data set, and then holds this resistance constant as other operating parameters,
refrigerants, and tube circuit configurations are computationally explored.
2.3.2.4 Operating Parameters 8 & 9: Subcool and Superheat
The number of evaporator tubes containing superheated-vapor and condenser tubes with
subcooled-liquid are controlled to one and two, respectively. It is important to limit the number
of tubes in superheat and subcooling for comparison with the CYCLE_D-HX model because the
20 25 30 35 40 1.2
1.4
1.6
1.8
2.0
2.2
5.1
6.0
6.8
7.6
8.5
9.4
f luid = R134a
1.6
1.8
2.0
2.2
2.4
2.6
2.8
5.6
6.3
7.0
7.7
8.4
9.1
9.8
f luid = R134a
19
model does not consider the pressure drop or area required for heat transfer in these sections,
rather, it only accounts for transport phenomena in the two-phase section. (Note that the
superheat and subcooling are included as an input to the model to determine the thermodynamic
states at the heat exchanger outlets.) The superheat and subcooling are also limited because of
performance considerations; the single phase fluid exhibits a lower heat transfer coefficient
compared to a two-phase fluid. The superheat is bound at the low end by the need to prevent
liquid from entering the compressor. The lower limit for subcooling is governed by a
requirement of fully subcooled liquid entering the expansion valve; vapor bubbles entering the
expansion valve cause oscillations in the flow that make it difficult to achieve steady state. In
general, a superheat range of 3 K to 6 K and a subcooling range of 2 K to 3 K satisfies this
Operating Parameter requirement.
Closing the refrigerant needle valve increases the superheat and the subcooling, whereas
adding refrigerant increases the subcooling and reduces the superheat. The combination of these
two controls allows for independent regulation of superheat and subcooling.
2.3.2.5 Operating Parameter 10: LLSL-HX included or bypassed
The LLSL-HX heat exchanger is included or bypassed using the six valves shown in Figure
2-1. When the LLSL-HX is bypassed, only one (rather than both) of the valves that stops flow
through the heat exchanger is closed; this prevents the possibility of trapping subcooled liquid
between the valves (and presenting the same pressure hazard discussed in Section 2.1.
20
3 Data Analysis
3.1 Thermodynamic States
The thermodynamic states numbered 1 through 11 in Figure 2-1 are fixed with two intensive
properties defined by the measurements and equations presented in Table 3-1. Thermodynamic
property data for R134a are computed using the Engineering Equation Solver (EES) software
(Klein, 2015); the software uses the equation of state developed by Tilner-Roth et al. (1994) for
R134a. Note that property data for mixtures and for fluids not directly available in EES will be
computed using the NIST REFPROP database (Lemmon et al., 2013).
Table 3-1: Measurements and equations used to define the thermodynamic states
State Description Measurement Equations
1 Compressor inlet, LLSL-HX vapor outlet T1 P1 = P11 – Ps
2 Compressor cylinder before compression T2 = T1 P2 = P1
3 Compressor outlet T3 P3 = P4
4 Condenser inlet T4 P4 = P7 + Pc
5 Condenser saturated vapor P5 = P4 x5 = 1
6 Condenser saturated liquid P6 = P7 x6 = 0
7 Condenser outlet, LLSL-HX liquid inlet T7, P7
8 Expansion valve inlet, LLSL-HX liquid outlet T8 P8 = P7
9 Evaporator inlet, Expansion valve outlet P9 h9 = h8
10 Evaporator saturated vapor P10 = P11 x10 = 1
11 Evaporator outlet, LLSL-HX vapor inlet T11 P11 = P9 – Pe
T = temperature, P = pressure, h = enthalpy, x = thermodynamic quality
3.2 Thermodynamic Performance
Given the fixed state points identified in Table 3-1 (and all associated intensive properties),
the key thermodynamic performance metrics are calculated according to the equations in
Sections 3.2.1 through 3.2.7. Nominal values and uncertainties of the metrics are shown in
Section 4, and Appendix A, respectively.
3.2.1 Capacity
according to:
( ), 4 7c ref refQ m h h= − (3.1)
( ), 9 11e ref refQ m h h= − (3.2)
c, ,c , , ,HTF HTF p HTF c c ins cQ m C T Q= − (3.3)
21
, , , , ,e HTF HTF e p HTF e e ins eQ m C T Q= − (3.4)
where the variables represent:
,eHTFm ,HTF cm = mass flow of HTF in evaporator/condenser
C p,HTF,e
C p,HTF,c
,ins eQ ,ins cQ = insulation heat leak in evaporator/condenser
Te Tc = temperature change of HTF in evaporator/condenser
The insulation heat leak is a non-trivial 30 W to 50 W, and therefore must be accounted for in the
heat exchanger capacity. The heat leak is debited from the HTF-side capacity because the HTF
is the annulus, and therefore thermally interacts with the surrounding ambient air. The heat
exchangers are divided into active and inactive sections (i.e., tubes where the HTF flows, but the
refrigerant does and does not, respectively) to compute the heat leak according to:
( ) ( ), e,
e active inactive
ins e ins e HTF e active avg amb e HTF e inactive avg amb e
e total total
NT NT
c active c inactive
ins c ins c HTF c active avg amb c HTF c inactive avg amb c
c total c total
NT NT
NTe,total NTe,total = number of tubes in evaporator/condenser total (20)
NTe,active NTc,active = number of active tubes in evaporator/condenser
NTe,inactive NTe,inactive = number of inactive tubes in evaporator/condenser
THTF,e,active,avg THTF,c,active,avg = avg. HTF temp. in the active evaporator/condenser tubes
THTF,e,inactive,avg THTF,c,inactive,avg = avg. HTF temp. in the inactive evaporator/condenser tubes
Tamb,e Tamb,c = temp. of ambient air surrounding evaporator/condenser
The HTF temperatures are taken from the surface-mounted thermocouples, and the ambient air
temperature is taken from a thermocouple near each of the heat exchangers. The heat exchangers
are split into active/inactive sections because the slope of the HTF temperature profile is
different in each section; the heat transfer is only governed by the average temperature difference
if the slopes (and specific heats) are constant throughout the section. The UAins,e and UAins,c
values for all 20 tubes are 5.48 W K-1 ±0.31 W K-1 and 6.52 W K-1 ±0.41 WK-1 (10.4 Btu h-1 °F-1
±0.59 Btu h-1 °F-1 and 12.4 Btu h-1 °F-1 ±0.78 Btu h-1 °F-1). The method used to compute the
conductance values and their uncertainties is described in Appendix A.8.
3.2.2 Compressor Power
The compressor power is computed using the change in enthalpy across the compressor:
22
( )3 1comp refW m h h= − (3.7)
The shaft power is computed for comparison with the compressor power:
2shaftW Nπτ= (3.8)
where the variables represent:
N = compressor shaft speed
τ = compressor shaft torque
3.2.3 Coefficient of Performance
The cooling and heating COP are computed using the refrigerant-side capacity and the
compressor power applied to the refrigerant:
,COPcool e ref compQ W= (3.9)
,COPheat c ref compQ W= (3.10)
3.2.4 Volumetric Capacity
The volumetric cooling and heating capacity (VCC and VHC) are:
( ), 1e ref ref VCC Q m v= (3.11)
( ), 1c ref ref VHC Q m v= (3.12)
where:
3.2.5 Compressor Efficiency
( )3 1 1
The volumetric efficiency is quantified by the refrigerant displacement normalized by the
compressor swept volume displacement:
3.2.6 Superheat and Subcooling
The evaporator superheat is:
and the condenser subcooling is:
( )7 7, 0SC T P x T= = − (3.16)
where:
T(P11, x = 1) = saturated vapor temperature at the evaporator exit pressure
T(P7, x = 0) = saturated liquid temperature at the condenser exit pressure
The number of refrigerant tubes containing superheated/subcooled fluid in the
evaporator/condenser is controlled to one and two, respectively, during experimental tests.
Visual inspection of the temperature profile, such as the one shown in Figure 3-1, is used to
determine the locations of the various phase transitions. The tube number on the x-axis reflects
the convention established in Figure 2-5.
(a) (b)
Figure 3-1: Temperature profile in (a) evaporator, showing superheat and (b)
condenser, showing subcooling
*The location 21 is the exit of the 20th tube
3.2.7 LLSL-HX Effectiveness
The LLSL-HX effectiveness is computed as the ratio of the heat transferred on the vapor
24
( )
C T TQ
− = =
−
where:
Cp,avg,1,11 = average specific heats of the refrigerant on the vapor side
Cp,avg,7,8 = average specific heats of the refrigerant on the liquid side
The heat transfer on the vapor side, rather than liquid side, is chosen for the numerator because
the larger temperature difference exhibited on the vapor side (due to lower specific heat) results
in a smaller uncertainty in the effectiveness.
25
4 Validation and Baseline Tests with R134a
An initial set of tests were carried out with R134a to provide verification of the
instrumentation and show the nominal uncertainties of the measurements. One of the most
important validations is to show an energy balance between the refrigerant and the HTF in the
heat exchangers. The energy imbalance, defined as ( ) ,ref ref HTF Q Q Q− for the initial tests is
shown in Figure 4-1; the imbalance is generally around 1 % to 1.5 % or better in both the
evaporator and condenser. The imbalance falls well within its uncertainty at the 95 %
confidence interval, indicating that the estimation of the uncertainty in HTF capacity (3 %) is
likely too conservative; note that the HTF capacity accounts for more than 90 % of the relative
uncertainty in the imbalance metric.
Figure 4-1: Energy imbalance in the evaporator and condenser
The performance of R134a in the MBBHP, as described by the metrics defined in Section
3.2, are presented in Figure 4-2 through Figure 4-4. Figure 4-2 (a) and (b) show nominal
pressure profiles on the low/high pressure sides of the cycle, where the error bars represent
measurement uncertainty listed in Table 2-3. The compressor power measured both by the shaft
power and change of enthalpy of the working fluid are shown in Figure 4-3(a), where the
conversion from mechanical to fluid power is about 65 % efficient. The error bars represent the
uncertainty calculated by propagating measurement uncertainty through Eqs. (3.7) and (3.8).
The mass flow ranges from 8 gs-1 to 13 gs-1 (1.1 lb min-1 to 1.7 lb min-1). Figure 4-4 shows the
COP ranges from 2.4 to 4, and the capacity ranges from 1.6 kW to 2.1 kW (0.45 ton to 0.60 ton).
The figure shows the uncertainties of the COP and capacity, where the computation and
tabulation of uncertainty is described further in Appendix A.10 and A.11. The volumetric
capacity ranges from 2100 kJ m-3 to 2800 kJ m-3 (56 Btu ft-3 to 75 Btu ft-3), and the compressor
isentropic and volumetric efficiencies are nominally 0.46 and 0.55; uncertainties in all three of
these metrics are shown in the figures and are computed using a similar method as the
capacity/COP, but the computation is not discussed in detail in this report.
Figure 4-3 (b) also shows the mass flux, which is used to determine the two-phase flow
regimes indicated in Figure 4-5. The refrigerant generally entered the annular flow regime in the
evaporator and condenser, though for some tests with low mass flux the refrigerant never left the
stratified-wavy regime.
(a) (b)
Figure 4-2: Heat pump cycle pressures on the (a) low pressure side and (b) high
pressure side
(a) (b)
Figure 4-3: Performance metrics including (a) compressor power and (b) mass flow and
mass flux
1100
1150
1200
350
400
450
0.6
0.9
1.2
shaft pow er (Wshaf t)
20 25 30 35 40
0.008
0.009
0.01
0.011
0.012
0.013
140
160
180
200
220
lo w
[ k g
lu x [
k g
/s -m
(a) (b)
(c) (d)
Figure 4-4: Performance metrics including (a) COP, (b) capacity, (c) volumetric
capacity, and (d) compressor isentropic and volumetric efficiency
20 25 30 35 40 2
2.5
3
3.5
4
4.5
1.6
1.7
1.8
1.9
2
2.1
2200
2400
2600
2800
0.5
0.6
0.7
0.3
0.4
0.5
0.6
0.7
(a)1,3 (b)2,3
Figure 4-5: Flow regime map for microfin tube (a) evaporation and (b) condensation
1 Flow boiling regime map from (Wojtan et al., 2005) 2 Flow condensation regime map from (Hajal et al., 2003) 3 Boundary between Intermittent/Annular regimes from (Thome, 2007), using equation 12.6.1 with a coefficient
of 0.678, for helical microfins
5 Conclusions and Future Work
This report shows that the test apparatus is capable of measuring the performance of
low-GWP refrigerants with high fidelity, including less than 0.5 % uncertainty in both capacity
and COP. The energy balance closure on the evaporator and condenser was better than 1.5 %
and will be continually tracked to ensure continued high quality data are collected.
Future tests will include a second baseline refrigerant R410A, followed by a series of
comparative tests with low-GWP refrigerants and refrigerant mixtures. The HTF flowrates will
remain fixed from the R134a baseline tests, so that the CYCLE_D-HX simulations of the test
data can be carried out for all test data based on a single reference data set. After the model
results are verified, the model will be used to identify additional candidate refrigerants to test in
the MBBHP test apparatus.
50
100
150
200
250
300
350
lu x [
k g
/s -m
M
S = Stratified SG = Slug SW = Stratified-Wavy
SW
50
100
150
200
250
300
350
lu x [
k g
/s -m
A
I
SG
Vapor quality A = Annular D = Dryout I = Intermittant M = Mist
S = Stratified SG = Slug SW = Stratified-Wavy
29
AHRI. (2008). 2008 Standard for Performance Rating of Unitary Air-Conditioning & Air-
Source Heat Pump Equipment. Air-Conditioning, Heating, and Refrigeration Institute,
Arlington, VA, United States. Retrieved from www.ahrinet.org
AHRI. (2015). Participants' Handbook: AHRI Low-GWP Alternative Refrigerants Evaluation
Program (Low-GWP AREP). Arlington, VA, United States: Air-Conditioning, Heating,
and Refrigeration Institute. Retrieved from
http://www.ahrinet.org/site/514/Resources/Research/AHRI-Low-GWP-Alternative-
Refrigerants-Evaluation
Brown, J. S., Domanski, P. A., & Lemmon, E. (2011). Standard Reference Database 49,
CYCLE_D: NIST Vapor Compression Cycle Design Program, Version 5.0. Retrieved
from http://www.nist.gov/srd/nist49.cfm
Domanski, P. A., Brown, S. J., Heo, J., Wojtusiak, J., & McLinden, M. O. (2014). A
thermodynamic analysis of refrigerants: Performance limits of the vapor compression
cycle. International Journal of Refrigeration, 38, 71-79.
doi:10.1016/j.ijrefrig.2013.09.036
EPA. (2015). 2015 North American Amendment Proposal to Address HFCs under the Montreal
Protocol. Retrieved from http://www.epa.gov/ozone/intpol/mpagreement.html and
http://www.epa.gov/ozone/intpol/HFC_Amendment_2015_Summary.pdf
EU. (2014, April 16). Regulation (EU) No 517/2014 of the European Parliament and of the
Council of 16 April 2014 on fluorinated greenhouse gases and repealing Regulation (EC)
No 842/2006. Journal of the European Union. Retrieved from http://eur-
lex.europa.eu/legal-content/EN/TXT/?uri=uriserv:OJ.L_.2014.150.01.0195.01.ENG
Fortin, T. J. (2015). Personal communication.
Hajal, J. E., Thome, J. R., & Cavallini, A. (2003). Condensation in horizontal tubes, part 1: two-
phase flow pattern map. International Journal of Heat and Mass Transfer, 3349-3363.
doi:10.1016/S0017-9310(03)00139-X
Harr, L., Gallagher, J. S., & Kell, G. S. (1984). NBS/NRC Steam Tables. Hemisphere Publishing
Co.
Kazakov, A., McLinden, M. O., & Frenkel, M. (2012). Computational Design of New
Refrigerant Fluids Based on Environmental, Safety, and Thermodynamic Characteristics.
Industrial & Engineering Chemistry Research, 51, 12537-12548. doi:10.1021/ie3016126
Kim, Y., Payne, V., Choi, J., & Domanski, P. A. (2005). Mass flow rate of R-410A through short
tubes working near the critical point. International Journal of Refrigeration, 28, 547-553. doi:10.1016/j.ijrefrig.2004.10.007
Klein, S. A. (2015). Engineering Equation Solver, v 9.908. Retrieved from
http://www.fchart.com/ees/
Lemmon, E. W., Huber, M. L., & McLinden, M. O. (2013). NIST Standard Reference Database
23, Reference Fluid Thermodynamic and Transport Properties (REFPROP), Version 9.1.
30
http://www.nist.gov/srd/nist23.cfm
McLinden, M. O., & Radermacher., R. (1987). Methods for Comparing the Performance of Pure
and Mixed Refrigerants in the Vapour Compression Cycle. International Journal of
Refrigeration, 10(6), 318-325. doi:10.1016/0140-7007(87)90117-4
McLinden, M. O., Kazakov, A., Heo, J., Brown, J. S., & Domanski, P. A. (2014). A
thermodynamic analysis of refrigerants: Possibilities and tradeoffs for Low-GWP
refrigerants. International Journal of Refrigeration, 38, 80-92.
doi:10.1016/j.ijrefrig.2013.09.032
Pannock, J., & Didion, D. A. (1991). The Performance of Chlorine-Free Binary Zeotropic
Refrigerant Mixtures in a Heat Pump, NISTIR 4748. Internal Report, National Institute of
Standards and Technology, U.S. Department of Commerce, Gaithersburg. Retrieved from
http://fire.nist.gov/bfrlpubs/build91/art002.html
Solomon, S., Qin, D., Manning, M., Chen, Z., Marquis, M., Averyt, K. B., . . . Miller, H. L.
(2007). Contribution of Working Group I to the Fourth Assessment Report of the
Intergovernmental Panel on Climate Change. New York, NY, USA: Cambridge
University Press. Retrieved July 29, 2015, from
http://www.ipcc.ch/publications_and_data/publications_ipcc_fourth_assessment_report_
wg1_report_the_physical_science_basis.htm
Taylor, B. N., & Kuyatt, C. E. (1994). Guidelines for Evaluating and Expressing the Uncertainty
of NIST Measurement Results, National Institute of Standards and Technology Technical
Note 1297.
Thome, J. R. (Updated in 2007). Wolverine Heat Transfer Engineering Data book III. Germany:
Weiland-Werke AG. Retrieved from http://www.wlv.com/heat-transfer-databook/
Tilner-Roth, R., & Baehr, H. D. (1994). An International Standard Formulation for the
Thermodynamic Properties of 1,1,1,2-Tetrafluoroethane (HFC-134a) for Temperatures
from 170 K to 455 K and Pressures up to 70 MPa. J. Phys. Chem, Ref. Data, 23(5).
Wojtan, L., Ursenbacher, T., & Thome, J. R. (2005). Investigation of flow boiling in horizontal
tubes: Part I—A new diabatic two-phase flow pattern map. International Journal of Heat
and Mass Transfer, 2955-2969. doi:10.1016/j.ijheatmasstransfer.2004.12.012
31
Symbol Units Definition
Q kW Energy transfer
SX -- Standard deviation of measured quantity “X”
T °C Temperature
TTC,cal °C Calibrated thermocouple measurement
TTC,CJC °C Thermocouple measurement with Cold Junction Compensation
Tlm K Log-mean temperature difference
TTC,cal K Thermocouple calibration temperature difference
TTP K Temperature difference of thermopile
UA W K-1 Thermal conductance
, ,UA ins e U W K-1 Uncertainty of average of evaporator insulation conductance
UTC °C Total instrument uncertainty of thermocouple
UTC,cal °C Thermocouple instrument uncertainty from calibrating device uncertainty
UTC,fit °C Instrument uncertainty of thermocouple due to fit of calibration regression
UX -- Uncertainty of measured quantity “X”
X U -- Uncertainty of average of measured quantity “X”
UY Uncertainty of calculated quantity “Y”
V V Voltage
V1 V Thermopile voltage of cold end of thermopile (referenced to ice-point)
VCC kJ m-3 Volumetric Cooling Capacity
VHC kJ m-3 Volumetric Heating Capacity
VTC V Thermocouple voltage
VTP V Thermopile voltage
in Inlet
ins Insulation
out Outlet
1 to 11 Refrigerant thermodynamic states as defined by Figure 2-1
Abbreviation Definition
COP Coefficient of Performance (W capacity per W of input power)
CJC Cold Junction Compensation (for thermocouples)
DAQ Data Acquisition
DSC Differential Scanning Calorimeter (used to measure fluid specific heat)
EES Engineering Equation Solver (software used to reduce data)
HTF Heat Transfer Fluid
RPM Revolutions per Minute (compressor shaft)
RTD Resistance Temperature Detector
A.2 General Remarks
The uncertainty analyses for key performance metrics are presented in this section. All
uncertainties are estimated based on instrument uncertainties computed at a 95 % confidence
interval.
The surface-mounted thermocouples, as well as the thermocouples measuring the air
temperature near the evaporator/condenser, are compensated at the DAQ using a thermistor-
based Cold Junction Compensation (CJC). A calibration (Tcal) is applied to the resulting
temperature reading (TTC,CJC) to compute the calibrated temperature (TTC,cal) according to:
,cal , ,TC TC CJC TC cal
T T T= + (A.1)
where the calibration is fit to a 2nd order polynomial (the order of this curve fit, and others
presented in Appendix A, were selected based on t-test of polynomial coefficients):
2
, , 0 1 , 2 ,TC cal cal TC CJC TC CJC TC CJCT T T a a T a T = − = + + (A.2)
The polynomial coefficients (a0, a1, a2) are fit using the least square error method, and Tcal is the
temperature measured by the calibrating instrument (two calibrated RTDs read by a precision
digital thermometer, expanded uncertainty of ±0.02 °C). The thermocouples were calibrated over
a temperature range of -10 °C to 60 °C, and this calibration was repeated five times over the
33
course of two weeks; Figure A-1 (a) shows the calibration data and polynomial curve fits for a
representative thermocouple. Each calibration shows a significant offset from the others; this is
likely a manifestation of the repeatability error of the CJC, which has an overall uncertainty of
±0.5 °C. Figure A-1 (b) shows the aggregation of the five data sets and the resulting curve fit
and 95 % confidence interval (±0.6 °C). The uncertainty of the calibrated RTDs is more than an
order of magnitude smaller than the curve fit confidence interval, and is therefore neglected. The
CJC thermocouple instrument uncertainty listed in Table 2-3 is therefore ±0.6 °C.
(a) (b)
Figure A-1: Temperature calibration for CJC compensated thermocouples where
(a) data are divided by each of the 5 calibrations and (b) all data are
combined.
A.4 Thermocouples with Ice-Water Bath Compensation
A lower level of uncertainty is achieved for the in-stream probe thermocouples, which
measure the refrigerant temperatures at key thermodynamic states, by utilizing an ice-water bath
reference junction. The ice-water bath consists of a vacuum insulated Dewar filled with crushed
ice and water; the reference junction is submerged in the bath inside of a water-filled test tube.
Using the ice-water bath removes the large uncertainty associated with the electronic CJC on the
DAQ system. The thermocouples were calibrated against the RTDs (same ones used to calibrate
thermocouples with CJC), where the resulting data and curve fit are shown for a representative
thermocouple in Figure A-2. The data were fit to 5th order polynomials according to:
5
=∑ (A.3)
where VTC is the thermocouple voltage. The resulting instrument uncertainty (±0.08 °C) listed in
Table 2-3 is the combination of the curve fit standard error (UTC,fit, k = 2) and the calibrating
RTD error (UT,cal):
2 2 2 2
, , 0.079 0.02 0.08TC TC fit T calU U U C= ± + = ± + = ± ° (A.4)
-10 0 10 20 30 40 50 60 -1.5
-1
-0.5
0
0.5
-1
-0.5
0
0.5
34
A.5 Thermopiles in Heat Transfer Fluid
The 13-junction thermopiles were calibrated by immersing one end in an ice-water bath
(inside a water-filled test tube) and the other end in a temperature controlled bath. The
temperature in the controlled bath was measured using the two calibrated RTDs. A thermopile
voltage-temperature dataset was generated and fit to a 4th order polynomial. Figure A-3 shows
the calibration data and the curve fit, and Table A-1 shows the polynomial coefficients. The
calibration data are fit to the equation:
4
The temperature difference for a thermopile is computed according to:
( )1 1TP TPT T V V T = + − (A.6)
where VTP is the measured thermopile voltage and T1 is the temperature measured at the cold
end of the thermopile by a CJC-corrected thermocouple. Eq. (A.5) is substituted into Eq. (A.6)
to calculate the thermopile temperature difference:
( ) 4
=
= + −∑ (A.7)
where V1 (cold-end voltage) is computed by numerically solving Eq. (A.5) where T is set to (T1):
4
=
= − =
∑ (A.8)
The thermopile uncertainties are computed with respect to the curve fit, voltage
measurement, and temperature measurement at the cold end of the thermopile (characterized
respectively by the coefficient uncertainties in Table A-1, the thermopile voltage uncertainty in
Table 2-3, and the CJC corrected thermocouple in Table 2-3). The EES (Klein, 2015) software
-0.0005 0 0.0005 0.001 0.0015 0.002 0.0025 -20
0
20
40
60
80
datadata
35
is used to numerically compute the partial derivatives and subsequent thermopile uncertainties.
Figure A-4(a) shows the contours of uncertainty in the temperature difference measured by the
evaporator/condenser HTF thermopiles.
Figure A-4(b) shows the contours of relative contribution to uncertainty in the thermopile
measurement. Each of the uncertainty sources is dominant in different regions:
• voltage measurement: dominant at low temperature differences (less than 1 K)
• curve fit error: dominant at high cold end temperatures (greater than 50 °C)
• cold end temperature: dominant at moderate cold end temperatures -5 °C to 20 °C and temperature differences 2.5 K to 20 K
As noted in this discussion, the thermopile uncertainty is a complex function of the curve fit,
voltage measurement, and cold end temperature measurement. For the temperature difference
range expected for this application, 1 K to 10 K (1.8 °F to 18 °F), the upper bound of uncertainty
is about 0.015 K (0.027 °F); for simplicity, this is the value listed in Table 2-3. Note that the
contribution to the uncertainty by the calibrating instrument (two calibrated RTDs) is negligible
here as the error in slope (which is the error that propagates into the thermopile temperature
difference) is less than 0.000248 K per K of temperature difference (e.g. 0.00248 K for a 10 K
temperature difference).
Table A-1: Condenser & evaporator
aTP,e,1 2.0243 E+03 ±1.2 E-01
aTP,e,2 -5.5466 E+03 ±2.2 E+01
aTP,e,3 3.9023 E+04 ±1.3 E+03
aTP,e,4 -2.6229 E+05 ±2.2 E+04
aTP,c,0 3.8494 E-02 ±5.3 E-04
aTP,c,1 2.0241 E+03 ±1.5 E-01
aTP,c,2 -5.5190 E+03 ±2.8 E+01
aTP,c,3 3.8551 E+04 ±1.6 E+03
aTP,c,4 -2.5440 E+05 ±2.9 E+04
-0.01 0 0.01 0.02 0.03 0.04
-10
0
10
20
30
40
50
60
(a) (b)
Figure A-4: Contours of thermopile (a) uncertainty and (b) relative contribution to
uncertainty
A.6 Evaporator Heat Transfer Fluid Specific Heat
The evaporator HTF is a potassium formate and water heat transfer fluid that is freeze
protected to -40 °C. The manufacturer’s specific heat data were found to have error of at least
10 %, so a sample of the fluid was analyzed in a Differential Scanning Calorimeter (DSC) by the
NIST-Boulder Applied Chemicals and Materials Division. The temperature dependent specific
heat and the 2nd order curve fit to the data are shown in Figure A-5; the curve fit coefficients and
their uncertainties are listed in Table A-2. Figure A-5 also shows the manufacturer’s reported
specific heat data for comparison. The exact uncertainty of the DSC measurement was not
available at the time of writing, but was estimated to be about 2 % to 3 % of reading based on the
experience of the DSC operator (Fortin, 2015). An uncertainty of 3 % of reading is therefore
assigned as a conservative estimate. Both the curve fit and DSC calibration uncertainties are
included in the calculation of overall uncertainties.
0 10 20 30 40 50 60 -5
0
5
10
15
0.035
0.002
0
5
10
15
20
T e m
p e ra
37
heat polynomial coefficients
aCp,e,1 1.1015 E-03 ±2.1 E-04
aCp,e,2 1.7422 E-05 ±9.5 E-06
Figure A-5: Specific heat measurements for
evaporator HTF
A.7 Condenser Heat Transfer Fluid Specific Heat
The condenser HTF is water; the specific heat data are computed in EES using the data from
(Harr et al., 1984) and are shown in Figure A-6. A conservative estimate of ±3 % is used for the
uncertainty in the specific heat.
Figure A-6: Specific heat of condenser HTF (water)
A.8 Evaporator and Condenser Insulation Conductance
The total evaporator and condenser insulation conductances (UAins,e and UAins,c) for thermal
interaction with the ambient air were measured during an initial test where the HTFs were
circulated at various flow rates through the apparatus at an elevated temperature, while the
refrigerant tubes were under vacuum. The temperature change of the HTF across the heat
-10 0 10 20 30
2.5
2.6
2.7
2.8
2.9
3
4.19
4.2
4.21
4.22
4.23
38
exchanger was assumed to be entirely attributed to insulation heat leak, and therefore an
effective insulation conductance could be computed. The resulting UAins,e and UAins,c values for
all 20 tubes are shown in Figure A-7. Average values of conductances for the evaporator and
condenser were calculated as 5.48 W K-1 ±0.31 W K-1 and 6.52 W K-1 ±0.41 W K-1
(10.4 Btu h-1 °F-1 ±0.59 Btu h-1 °F-1 and 12.4 Btu h-1 °F-1 ±0.78 Btu h-1 °F-1), respectively, where the
uncertainty accounts for both instrumentation error and test variance. The evaporator insulation
conductance for these tests is computed according to:
( )
( ) ( ) ( ) ( )
, , , , , , ,,
,
ln
HTF e p HTF e HTF in e HTF out eins e
ins e
lm HTF HTF in e amb e HTF e amb e
HTF in e amb e
HTF out e amb e
m C T TQ UA
T T T T T
T T
T T
,HTF e m = evaporator HTF mass flow
,ins eQ = evaporator insulation heat leak
Tamb,e = ambient air temperature surrounding evaporator
THTF,in,e = evaporator HTF inlet temperature
THTF,out,e = evaporator HTF outlet temperature
Tlm,HTF,e = evaporator log-mean temperature difference
and the uncertainty of the average value is calculated according to:
, ,, , UA ins eUA ins e U k S n= (A.10)
where the variables represent:
k = expanded uncertainty coverage factor (k = 2) for 95 % confidence interval
SUA,ins,e = standard deviation of conductances
n = number of conductance data points
The condenser conductance is computed using the same method with the corresponding
measurements from that heat exchanger.
39
A.9 Uncertainty Analysis Software
The uncertainty analyses on each of the performance metrics shown in Figure 4-1 through
Figure 4-4 were calculated using the uncertainty propagation capability in EES. The software
numerically computes the performance metric uncertainty by evaluating the partial derivatives
with respect to each relevant measurement and applying the specified measurement uncertainty.
Assuming the measurements are independent and uncorrelated, the uncertainty of a quantity is
computed according to (Taylor & Kuyatt, 1994):
2
2
Xi = measurement
5
6
7
8
A.10 Moving Window and Steady State Uncertainty
A moving window average was used to compute the measurements at steady state during the
tests. In order to determine a proper window size, the variation of measurement steady state
uncertainty with window size was examined. The steady state uncertainty is defined as the
expanded standard uncertainty for the average reading within the window:
X
X
n = number of samples in the window
Sx = standard deviation of measurement
X = average value of measurement
The data for analysis were generated by holding the MBBHP at a nominal steady operating
condition and recording the measurements at 1-minute intervals (same as collection rate for all
performance testing) for an extended time (3 hours). Figure A-8 shows the fluctuation of steady
state uncertainty for the evaporator inlet and outlet temperatures (T9 and T11). The figure
highlights two characteristic trends observed for all of the instruments, both of which suggest the
need for a large moving window: (1) the steady state uncertainty decreases as the window size
increases, and (2) the steady state uncertainty is highly oscillatory for small windows (the
oscillations are undesirable because short-term fluctuations can have a disproportionate impact
on the averaged result). A 30 sample moving window was selected to achieve low uncertainty
while allowing for collection of data in a timely manner.
Figure A-8 shows that the measurement steady state uncertainty is not constant in time (it
would be constant if the fluctuations at steady state were purely random), but rather exhibits
drifting and periodic movements. This is possibly due to hunting by the PID controllers
regulating HTF fluid temperatures, changes in air temperature in the lab space, or some other
uncontrolled and/or uncharacterized phenomena. These non-random fluctuations are accounted
for by using the maximum steady state uncertainty value observed during the extended steady
state test. For example, Figure A-8(a) shows the maximum steady state uncertainty for the
evaporator outlet (T9) is ±0.06 °C; this value and values for all other sensors are listed in Table
2-3. The data in the moving window are only recorded when steady state uncertainties are below
these established thresholds. The total uncertainty (±0.09 °C), also shown in Table 2-3, is
subsequently computed by adding the instrument uncertainty (±0.08 °C) and steady state
uncertainty (±0.06 °C) in quadrature.
Note that different sensors of the same type exhibit different levels of fluctuation; as a
conservative estimate, the largest uncertainty of any one is applied to all of them. For example,
Figure A-8(b) shows the fluctuations of steady state uncertainty for the evaporator inlet
temperature (T9) are much smaller than for the evaporator outlet (T11). In fact, the evaporator
outlet temperature (T11) shows the most fluctuation of all the ice-water bath compensated
thermocouple probes; therefore, its value (±0.06 °C) is used in Table 2-3 to represent them all.
41
(a) (b)
Figure A-8: Uncertainty in temperature due to fluctuations at steady state for (a)
evaporator outlet and (b) evaporator inlet
A.11 Uncertainty Results Summary
refrigerant/HTF-side capacities and the COPs are tabulated in Appendices A.11.1 (cooling) and
A.11.2 (heating). The tables show the performance metric and its total uncertainty (including
both instrument and steady state contributions), as well as the measurements that contribute to
the uncertainty. The measurement values, uncertainties, partial derivatives (of performance
metric with respect to the measurement), and relative errors (contribution to the overall error in
the performance metric) are also shown, where the measurements are listed in order of
decreasing relative error. Uncertainties for other performance metrics (e.g. VCC/VHC, ηisen)
were analyzed using similar techniques, but are not shown in detail in this report. Note that
Appendices A.11.1 and A.11.2 show sample calculations for a single nominal operating
condition, whereas the uncertainty values presented in the Section 4 figures are computed for
each test run.
The uncertainty in cooling performance metrics are shown for a nominal operating point for
R134a during a Cooling A test with a compressor speed of about 30 Hz (1800 RPM). Similarly,
the uncertainty in heating performance metrics are shown for R134a during a Heating H1 test
with a compressor speed of about 30 Hz (1800 RPM).
A.11.1 Uncertainty Results Summary: Cooling
Table A-3 shows that the dominant sources of instrument uncertainty for the refrigerant-side
capacity are refrigerant mass flow (66 %), expansion valve inlet temperature (17 %), evaporator
exit pressure (8 %), and evaporator exit temperature (7 %). Table A-4 shows the dominant
source of instrument uncertainty for the HTF-side capacity is the HTF specific heat (97 %,
estimate for instrument used to measure fluid properties). Table A-5 shows the dominant
sources of instrument uncertainty for cooling COP are evaporator inlet pressure (39 %),
compressor discharge temperature (25 %), and compressor suction temperature (18 %).
8000 12000 16000 20000 0.002
0.01
0.1
0.5
0.01
0.1
0.5
Parameter Symbol Value
(0.20%) -- --
Refrigerant mass flow refm 11.00 g s-1 ±0.024 g s-1 1.538 E-01 66 %
Refrigerant temp. at expansion valve inlet T8 38.16 C ±0.09 °C -1.634 E-02 17 %
Refrigerant press. at evaporator inlet P9 412 kPa ±3.5 kPa -2.817 E-04 8 %
Refrigerant temp. at evaporator exit T11 12.09 C ±0.09 °C 1.009 E-02 7 %
Refrigerant press. at condenser exit P7 1104 kPa ±3.5 kPa 2.103 E-06 0 %
Refrigerant evaporator differential press. Pe 43.92 kPa ±0.8 kPa 2.817 E-04 0 %
Table A-4: Uncertainty of HTF-side evaporator cooling capacity
Parameter Symbol Value
(3.0%) -- --
Evaporator HTF capacity
calibration uncertainty ,HTF ecp 2.59 J g-1 K-1 ±3 % of ,HTF ecp 6.369 E-01 97 %
Temperature of ambient air
surrounding evaporator Tamb,e 23.0 °C ±0.6 °C 5.474 E-03 1 %
HTF mass flow ,HTF em 131.2 g s-1 ±0.26 g s-1 1.257 E-02 1 %
Evaporator HTF thermopile voltage VTP,e 2.512 mV ±0.007 mV 6.537 E-01 1 %
Evaporator HTF thermopile
aTP,e,1 2.0243 E+03 ±1.2 E-01 8.832 E-04 0 %
aTP,e,2 -5.5466 E+03 ±2.2 E+01 1.559 E-05 0 %
aTP,e,3 3.9023 E+04 ±1.3 E+03 2.106 E-07 0 %
aTP,e,4 -2.6229 E+05 ±2.2 E+04 2.554 E-09 0 %
Evaporator HTF capacity
Temperature at cold end of
evaporator HTF thermopile THTF,e,out 15.42 °C ±0.6 °C -2.963 E-03 0 %
Evaporator insulation conductance UAins,e 5.474 W K-1 ±0.3052 W K-1 4.175 E-03 0 %
43
Parameter Symbol Value
±0.01
(0.35 %) -- --
Refrigerant press. at evaporator inlet P9 412 kPa ±3.5 kPa -1.232 E-02 39 %
Refrigerant temp. at compressor discharge T3 81.97 °C ±0.09 °C 5.571 E-02 25 %
Refrigerant temp. at compressor suction T1 12.73 °C ±0.09 °C 4.788 E-02 18 %
Refrigerant press. at condenser exit P7 1104 kPa ±3.5 kPa 5.144 E-03 7 %
Refrigerant temp. at expansion valve inlet T8 38.16 °C ±0.09 °C -2.746 E-02 6 %
Refrigerant temp. at evaporator exit T11 12.09 °C ±0.09 °C 1.696 E-02 2 %
Refrigerant evaporator diff. press. Pe 43.92 kPa ±0.8 kPa 7.432 E-04 2 %
Refrigerant condenser diff. press. Pc 32.34 kPa ±1.3 kPa 4.731 E-04 1 %
Refrigerant suction line diff. press. Ps 14.22 kPa ±0.3 kPa 1.314 E-03 0 %
Refrigerant mass flow refm 11.00 g s-1 ±0.016 g s-1 2.209 E-04 0 %
A.11.2 Uncertainty Results Summary: Heating
Table A-6 shows the dominant sources of instrument uncertainty for the refrigerant-side
capacity are refrigerant mass flow (85 %), condenser exit temperature (9 %), and condenser inlet
temperature (5 %). Table A-7 shows the dominant source of instrument uncertainty for the HTF-
side capacity is the HTF specific heat (97 %, estimate for instrument used to measure specific
heat of water in the reference). Table A-8 shows the dominant sources of instrument uncertainty
for cooling COP are evaporator inlet pressure (32 %), compressor discharge temperature (31 %),
and compressor suction temperature (22 %).
44
Parameter Symbol Value
(0.2%) -- --
Refrigerant mass flow refm 9.06 g s-1 ±0.016 g s-1 2.067 E-01 85 %
Refrigerant temp. at condenser exit T7 37.39 °C ±0.09 °C -1.343 E-02 9 %
Refrigerant temp. at condenser inlet T4 77.22 °C ±0.09 °C 9.443 E-03 5 %
Refrigerant press. at condenser exit P7 992.8 kPa ±3.5 kPa -1.296 E-04 1 %
Refrigerant condenser differential press. Pc 27.2 kPa ±1.3 kPa -9.028 E-04 0 %
Table A-7: Uncertainty of HTF-side condenser heating capacity
Parameter Symbol Value
(2.9%) -- --
Condenser HTF capacity calibration
uncertainty ,cHTFcp 4.183 J g-1 K-1 ±3 % of ,cHTFcp 4.262 E-01 97 %
Condenser insulation conductance UAins,c 6.521 W K-1 ±0.4071 W K-1 1.056 E-02 1 %
HTF mass flow ,cHTFm 98.81 g s-1 ±0.23 g s-1 1.804 E-02 1 %
Condenser HTF thermopile voltage VTP,c 2.318